Download Structural Investigations of Liposomes: Effect of Phospholipid

Structural Investigations of Liposomes:
Effect of Phospholipid Hydrocarbon Length and the
Incorporation of Sphingomyelin
A Thesis Submitted to the College of Graduate Studies and Research in
Partial Fulfillment of the Requirements for the
Degree of Master of Science in the
Department of Food and Bioproduct Sciences
University of Saskatchewan
By
Hayley Rutherford
2011
© Copyright Hayley Rutherford 2011
All Rights Reserved.
PERMISSION TO USE
In presenting this thesis in partial fulfillment of the requirements for a Postgraduate
degree from the University of Saskatchewan, I agree that the Libraries of this University may
make it freely available for inspection. I further agree that permission for copying of this thesis
in any manner, in whole or in part, for scholarly purposes may be granted by the professor or
professors who supervised my thesis work or, in their absence, by the Head of the Department or
the Dean of the College in which my thesis work was done. It is understood that any copying or
publication or use of this thesis or parts thereof for financial gain shall not be allowed without
my written permission. It is also understood that due recognition shall be given to me and to the
University of Saskatchewan in any scholarly use which may be made of any material in my
thesis.
Requests for permission to copy or to make other use of material in this thesis in whole or
part should be addressed to:
Head of the Department of Food and Bioproduct Sciences
University of Saskatchewan
Saskatoon, Saskatchewan
S7N 5A8
i
ABSTRACT
The liquid crystal morphologies of symmetrical diacyl phosphatidylcholine liposomes
examined in this research were found to be dependent on saturated hydrocarbon chain length.
Both
powder
x-ray
diffraction
and
synchrotron
mid-IR
microscopy
indicate
that
phosphatidylcholines with short hydrocarbon tails (i.e. ten and twelve carbons) were more likely
to form unilamellar liposomes while those with long hydrocarbon tails (i.e. eighteen and twenty
carbons) are more likely to form multilamellar liposomes. Hydrocarbon chain lengths of fourteen
and sixteen represented a transitional zone between these two liquid crystal morphologies. The
Fourier transform infrared (FTIR) spectra where a shoulder developed on the peak at
wavenumber 1750 cm-1 particularly highlights the change in the packing of adjacent molecules
in the transitional zone. Due to structural similarities between phosphatidylcholine and
sphingomyelin, further investigations were conducted on liposomes formed of a combination of
these compounds. Sphingolipid containing liposomes have the potential to enhance anti-cancer
drug delivery systems due to possible syngergistic effects between bioactive sphingolipids and
the drug itself. Using the gentle hydration method, liposomes composed of different ratios of 1,2distearoyl-sn-glycero-3-phosphocholine
(PC)
and
N-(tricosanoyl)-sphing-4-enine-1-
phosphocholine (SM) were formed. Based on synchrotron mid-IR microscopy, the addition of
SM to a PC liposome at ratios of 1PC:1SM, 2PC:1SM and 4PC:1SM variations in the packing of
adjacent molecules were detected but were not found to impact the ultimate supramolecular
structure of the mixture. Slight variations in packing are the result of the enhanced hydrogen
bonding capability of the SM headgroups and differences in hydrocarbon chain length.
ii
ACKNOWLEDGEMENTS
Dr. Michael Rogers, thank you for your guidance and confidence. I wish you, and the
growing Rogers family, all the best as you start your next journey. Dr. Nicholas Low, thank you
for your calm and steady presence. Many thanks to my advisory committee for their time and
expertise.
To the graduate students of the FABS department, thanks for all your help, in school and
out, and best of luck.
To the Canadian Dairy Commission, I am so grateful for your financial support and the
opportunities with which it has provided me.
iii
DEDICATION
This thesis is dedicated to all the wonderful humans who have helped me keep my head
up over the last two years. Much love and thanks to my beautiful family, Bacon Nation, groovy
Guelph cats, Ridgeway ladies and my stellar Saskatoon support network. I am lucky to have each
and every one of you in my life.
iv
TABLE OF CONTENTS
PERMISSION TO USE................................................................................................................... i
ABSTRACT.................................................................................................................................... ii
ACKNOWLEDGEMENTS........................................................................................................... iii
DEDICATION............................................................................................................................... iv
TABLE OF CONTENTS................................................................................................................ v
LIST OF FIGURES ....................................................................................................................... vi
LISTOF TABLES.......................................................................................................................... vi
LIST OF ABBREVIATIONS....................................................................................................... vii
1. INTRODUCTION ...................................................................................................................... 1
2. LITERATURE REVIEW ........................................................................................................... 3
2.1 Phospholipids........................................................................................................................ 3
2.1.1 Phosphatidylcholine....................................................................................................... 4
2.2 Sphingolipids ........................................................................................................................ 5
2.2.1 Sphingomyelin ............................................................................................................... 8
2.2.2 Sphingolipid Metabolism in Cancer Pathogenesis and Treatment ................................ 8
2.3 Supramolecular Strucures of Amphiphillic Lipids ............................................................... 9
2.3.1 Liposomes .................................................................................................................... 12
3. DEPENDENCE OF LIQUID CRYSTAL MORPHOLOGY ON PHOSPHOLIPID
HYDROCARBON LENGTH....................................................................................................... 15
3.1 Abstract ............................................................................................................................... 15
3.2 Objective and Hypothesis ................................................................................................... 15
3.3 Introduction......................................................................................................................... 15
3.4 Materials and Methods........................................................................................................ 17
3.5 Results and Discussion ....................................................................................................... 19
3.6 Conclusions......................................................................................................................... 28
3.7. Rationale Linking Research Studies .................................................................................. 29
4. STRUCTURE OF SPHINGOMYELIN-PHOSPHATIDLYCHOLINE LIPOSOMES........... 30
4.1 Abstract ............................................................................................................................... 30
4.2 Objective and Hypothesis ................................................................................................... 30
4.3 Introduction......................................................................................................................... 30
4.4 Materials and Methods........................................................................................................ 33
4.5 Results and Discussion ....................................................................................................... 35
4.6 Conclusion .......................................................................................................................... 41
5. GENERAL DISCUSSION ....................................................................................................... 42
6. GENERAL CONCLUSIONS................................................................................................... 45
7. FUTURE DIRECTIONS .......................................................................................................... 47
8. REFERENCES ......................................................................................................................... 48
v
LIST OF FIGURES
Figure 2.1 Selected pathways in phospholipid metabolism........................................................... 4
Figure 2.2 Representative chemical structure of phosphatidylcholine. ......................................... 5
Figure 2.3 Represenative chemical structures of sphingomyelin (A), ceramide (B), sphingosine
(C) and sphingosine-1-phosphate (D). .................................................................................. 6
Figure 2.4 Selected pathways in sphingolipid metabolism. Adapted from Ogretman and Hannun
(2004). ................................................................................................................................... 7
Figure 2.5 Critical packing parameters for amphiphillic molecules and the resulting critical
packing shapes and resulting liquid crystal structures. Adapted from Israelachvili (1991).
............................................................................................................................................. 11
Figure 3.1 Light micrographs of vesicle suspensions composed of PC10 (A), PC12 (B), PC14
(C), PC16 (D), PC18 (E), and PC20 (F). Scale bar indicates 5 µm. ................................... 20
Figure 3.2 Volume-moment mean diameters of phosphatidylcholine vesicles calculated using
pixel measurements obtained from light micrographs. ....................................................... 21
Figure 3.3 Light scattering intensities of phosphatidylcholine vesicle suspensions with varying
hydrocarbon chain lengths (A) PC10, (B) PC12, (C) PC14, (D) PC16, (E) PC18 and (F)
PC20. ................................................................................................................................... 22
Figure 3.4 Volume-moment mean diameters and surface area of phosphatidylcholine vesicles
calculated from dynamic light scattering measurements. Data points represent mean values
± one standard deviation. .................................................................................................... 23
Figure 3.5 Powder X-ray diffraction data for PC18 (A) and PC12 (B). ...................................... 24
Figure 3.6 FTIR spectra for liposomal suspensions of PC 10 (bottom spectrum) to PC 20 (top
spectrum) vesicle suspensions in water (A) and deuterated water (B)................................ 25
Figure 3.7 Oscillatory frequency measurements within the linear visco-elastic region for PC10
(A), PC12 (B), PC14 (C), PC16 (D), PC18 (E) and PC20 (F). ........................................... 27
Figure 4.1 Schematic representation of sphingolipid metabolism and their signaling activities,
adapted from Vesper, et al. (1999). ..................................................................................... 32
Figure 4.2 Light micrographs of liposomal suspensions composed of PC18 (A), SM (B),
1PC:1SM (C), 2PC:1SM (D) and 4PC:1SM (E) (Scale bar = 5 µm).................................. 35
Figure 4.3 Volume-moment mean diameters of liposomal suspensions calculated using pixel
measurements obtained from light micrographs. ................................................................ 37
Figure 4.4 Light scattering intensities of liposomal suspensions of PC18 (A), SM (B), 1PC:1SM
(C), 2PC:1SM (D) and 4PC:1SM (E).................................................................................. 38
Figure 4.5 Volume-moment mean diameters of liposomal suspensions calculated from dynamic
light scattering measurements. Data points represent mean values ± one standard deviation.
............................................................................................................................................. 39
Figure 4.6 FTIR spectra for suspensions composed of PC, SM, 1PC:1SM, 2PC:1SM and
4PC:1SM. ............................................................................................................................ 40
LIST OF TABLES
Table 3.1 Power law coefficients of the storage (G! =a"b) and loss modulus (G# =c"d) of
liposomal suspensions compared with those for typical materials as determined by Steffe
(1996). ................................................................................................................................. 28
vi
LIST OF ABBREVIATIONS
EPR
Enhanced permeability and retention
FTIR
Fourier transform infrared
G!
Storage modulus
G"
Loss modulus
PC
Phosphatidylcholine
PC 10
1,2-didecanoyl-sn-glycero-3-phosphocholine
PC 12
1,2-dilauroyl-sn-glycero-3-phosphocholine
PC 14
1,2-dimyristoyl-sn-glycero-3-phosphocholine
PC 16
1,2-dipalmitoyl-sn-glycero-3-phosphocholine
PC 18
1,2-distearoyl-sn-glycero-3-phosphocholine
PC 20
1,2-diarachidoyl-sn-glycero-3-phosphocholine
PE
Phosphatidylethanolamine
PEG
Polyethylene glycol
PI
Phosphatidylinositol
PS
Phosphatidylserine
SM
Sphingomyelin
vii
1. INTRODUCTION
In aqueous solution the contrasting effects of the hydrophobic and hydrophilic moieties
of amphiphilic molecules cause the molecules to assemble into supramolecular liquid crystal
structures (Tanford 1980). In the case of phospholipids, the hydrophilic head groups associate
with the aqueous phase whereas the hydrophobic hydrocarbon chains interact with each other.
Characteristics of the particular phospholipid, or combination of phospholipids, also contribute
to the type of supramolecular structure. For example, the charge of head group can contribute to
ionic repulsion of adjacent molecules (Israelachvili 1991).
Lipsomes (i.e. phospholipid bilayer vesicles) have been the subject of investigation since
the early 1960’s. The potential for liposomes to encapsulate both water and lipid soluble
compounds has also been explored since the advent of liposome research (Bangham 1980;
Gregoriadis 1980; Drummond et al. 1999). In the initial investigations as well as current
research, multiple bilayer structures have been observed (Bangham 1980; Gregoriadis 1980). As
research in the area of liposomes began to focus on using them as mimics of a cell membrane,
unilamellar liposomes became the standard system for observation (Bangham 1980; Gregoriadis
1980; Drummond et al. 1999). Unilamellar liposomes also represent the standard for liposomes
prepared for research into drug delivery capabilities of liposomes, thus many researchers employ
methods to reduce lamellarity. Membrane extrusion and microfluidization are both employed to
ensure the homogeneity of liposomal suspensions.
Characteristics of both the hydrophobic hydrocarbon tails and hydrophilic head groups of
the phosopholipid can be used to predict which structures will form (Israelachvili 1991). These
parameters most efficiently predict the structures of unilamellar, single component
supramolecular structures (Israelachvili 1991). In a multilamellar structure the interaction of
head groups between separate lamellae, the behaviour of the aqueous solution within the
multilamellar system and the interaction of adjacent molecules within in each lamellae must all
be considered. The inclusion of multiple phospholipid components in a single supramolecular
1
structure provides an opportunity to investigate the effects of head group charge, hydrocarbon
chain length and hydrocarbon saturation.
Liposomes have been widely investigated as drug delivery systems, particularly for the
delivery of anti-cancer drugs (Bangham 1980; Gregoriadis 1980; Drummond et al. 1999).
Liposomes that have been sterically stabilized with compounds such as polyethylene glycol
(PEG) have contributed to increased therapeutic indices of anti-cancer and anti-fungal drugs
(Drummond et al. 1999). This is the result of increased circulation times and reduced toxicity
compared to the free drug (Allen 1996; Johnston, Semple et al. 2006). The versatility of
liposomes extends to the encapsulation of both lipid and water soluble drugs as well as the
inclusion of bioactive amphiphilic lipids into the structure itself.
Sphingolipids are a class of bioactive amphiphilic lipids that play roles in the cell
signaling pathways associated with senescence and apoptosis (i.e. programmed cell death)
(Ogretmen and Hannun 2004; Hannun and Obeid 2008). Sphingomyelin, sphingosine and
ceramide belong to this class. It is possible that this function can be exploited to increase the rate
of cell death in cancer treatment. Supramolecular structures of phosphatidylcholine (PC) and
sphingomyelin (SM) were investigated. Both PC and SM possess a phosphocholine head group
and two hydrocarbon chains, there are differences in their structure that in turn effect their
behaviour in phospholipid bilayer systems (Terova et al. 2004). A drug delivery liposome that
contains sphingolipids, namely SM, has the potential to create a drug delivery system that can
combine the effect of liposomally encapsulated anti-cancer drugs with the anti-cancer effects of
the released sphingolipids. Investigations of their compatibility in liposomal structures must
precede further research.
2
2. LITERATURE REVIEW
2.1 Phospholipids
Phospholipids are a class of amphiphilic lipids characterized by their negatively charged
phosphate head group (at physiological pH) and non-polar fatty acid components (Kent 1995;
Drummond et al. 1999; Li and Vance 2008). Reference to phospholipids in this research
particularly means diacyl-glycerophospholipids which are composed of a phosphate head group,
a diacylglycerol, and an additional simple organic molecule as part of the head group structure.
The addition of organic molecules to the phosphate head group creates a variety of phospholipid
species such as phosphotidylethanolamine (PE), phosphotidylserine (PS), phosphatidylinositol
(PI) and phosphatidylcholine (PC) (Kent 1995). Phospholipids play structural roles in cell
membranes and are capable of orienting in a variety of supramolecular structures in aqueous
solution (Israelachvili 1991; Kent 1995; Li and Vance 2008). The formation of supramolecular
structures in aqueous solution is driven by the hydrophobic interactions of the hydrocarbon
chains (Tanford 1980; Israelachvili 1991; Kent 1995).
The biosynthetic precursor for the diacyl-glycerophospholipids is phosphatidate (i.e. 1,2dicacyl-sn-glycero-3-phosphate). The pathway described in Figure 2.1 occurs in mammalian
cells (Kent 1995; Li and Vance 2008). Phosphatidate is synthesized de novo from glycerol-3phosphate, a product of glycolysis, and two acyl-CoA molecules via the action of glycerol-3phosphate acyltransferase. There are two forms of this enzyme, one catalyzing the reaction at
carbon-1 and the other at carbon-2 (Kent 1995; Li and Vance 2008).
A divergence in the synthesis pathway occurs here: 3-sn-phosphatidate phosphohydrolase
may cleave the phosphate group from carbon-3 leaving a diacylglycerol (DAG) or cytidine
diphosphate-diacylglycerol (CDP-DAG) may be formed by the action of CDP-DAG synthase
which catalyzes the addition of CDP to the existing phosphate group from cytidine triphosphate
(CTP) (Kent 1995; Li and Vance 2008). The DAG branch leads to the production of PC, PE and
PS while the CDP-DAG branch leads to the production of PI, PS, and other lipid molecules. This
3
research utilized PC, as many previous studies have to investigate phospholipid structures (Kent
1995; Li and Vance 2008).
Figure 2.1 Selected pathways in phospholipid metabolism.
2.1.1 Phosphatidylcholine
Phosphatidylcholine (Figure 2.2) is abundant in both plant and animal cell membranes; it
may account for 50-90% of cell membrane phospholipids and is largely located in the outer
leaflet of the cell membrane (Kent 1995; Li and Vance 2008). It has been used as a food additive,
emulsifier and in the liposomal encapsulation of flavours, textile dyes and pharmaceuticals
(Drummond et al. 1999). Phosphatidylcholine in mammals is primarily synthesized from the
DAG branch of the phospholipid synthetic pathway. Phosphocholine is transferred to carbon-3
via the action of choline:1,2-diacylglycerol cholinephosphotransferase (Kent 1995; Li and Vance
2008).
4
Figure 2.2 Representative chemical structure of phosphatidylcholine.
2.2 Sphingolipids
Sphingolipids are a class of lipids that contain a sphingoid base; an amino-alcohol
molecule. These compounds share their amphiphillic characteristics with and have similar
structural roles in the cell membrane as phospholipids (Ogretmen and Hannun 2004; Hirabayashi
et al. 2006; Hannun and Obeid 2008). Typically, they are concentrated in microdomains of the
cell membranes called lipid rafts. These domains have a higher concentration of sphingolipids,
cholesterol and membrane bound lipoproteins than the surrounding cell membrane. These raft
structures play an important role in cell-cell recognition, ligand interaction and cell attachment
(Ogretmen and Hannun 2004; Hirabayashi et al. 2006; Hannun and Obeid 2008). This
understanding of membrane structure led to further investigations of the nature of sphingolipids
and revealed their importance as signaling molecules when previously they had been thought to
5
serve only a structural role (Ogretmen and Hannun 2004; Hirabayashi et al. 2006; Hannun and
Obeid 2008).
Sphingolipids are associated with the signaling pathways that influence cell proliferation,
senescence, inflammatory responses and apoptosis. Effector molecules from the sphingolipid
family include ceramide, sphingosine and sphingosine-1-phosphate. (Ogretmen and Hannun
2004; Hirabayashi et al. 2006; Hannun and Obeid 2008)
A
B
C
D
Figure 2.3 Representative chemical structures of sphingomyelin (A), ceramide (B), sphingosine
(C) and sphingosine-1-phosphate (D).
6
Sphingolipid metabolism is highly diverse and its full complexity has yet to be entirely
elucidated. Ceramide is considered to be a central molecule in sphingolipid metabolism because
of its role in an array of anabolic and catabolic reactions (Ogretmen and Hannun 2004). Figure
2.4 illustrates some selected pathways in sphingolipid metabolism.
Figure 2.4 Selected pathways in sphingolipid metabolism. Adapted from Ogretman and Hannun
(2004).
7
2.2.1 Sphingomyelin
Sphingomyelin (SM) is the most common sphingolipid in mammalian cell membranes.
Like PC, it is typically located on the outer leaflet of the cell membrane(Niemela et al. 2004;
Hirabayashi et al. 2006). Also as with PC, the fatty acid composition of SM in mammalian cells
differs based on location in the body. Sphingomyelin is synthesized de novo from
phosphatidylcholine and ceramide according to the following reaction catalyzed by
sphingomyelin synthase:
ceramide + phosphatidylcholine ! sphingomyelin + 1,2-diacyl-sn-glycerol
A pathway for ceramide synthesis is the cleavage of the phosphocholine group of SM by
sphingomyelinase (Ogretmen and Hannun 2004; Hirabayashi et al. 2006; Hannun and Obeid
2008). Since SM represents a large portion of the sphingolipids found in mammalian cells, it is a
significant source for the production of ceramide from the cell membrane for signaling purposes
(Ogretmen and Hannun 2004; Hirabayashi et al. 2006; Hannun and Obeid 2008).
2.2.2 Sphingolipid Metabolism in Cancer Pathogenesis and Treatment
A number of sphingolipids have been shown to be of significant importance in cell
signaling pathways associated with senescence and apoptosis (Ogretmen and Hannun 2004;
Hannun and Obeid 2008). Apoptosis is the collection of biochemical signaling pathways that
lead to programmed cell death (Ogretmen and Hannun 2004; Hannun and Obeid 2008). Cancer
pathogenesis involves disruptions in the apoptitic pathway (Ogretmen and Hannun 2004;
Hannun and Obeid 2008). Without the regulation of cell death, cancer cells have an increased
capacity to proliferate and induce angiogenesis (i.e. the formation of new blood vessels)
(McDonald and Baluk 2002). In a normal cell, ceramide is involved in the regulation of apotosis
as well as senescence and growth arrest. Ceramides have specific enzymatic targets, such as
ceramide-activated protein phosphatases and D-cathespin, that in turn participate in cascades that
have the downstream effects of cell growth arrest, senescence and apoptosis (Ogretmen and
Hannun 2004; Hannun and Obeid 2008).
Atypical levels of sphingolipids and enzymes involved with their metabolism, compared
with normal cell metabolism, have been noted in different stages of cancer development (Hannun
and Obeid 2008). The application of this knowledge has led to research in the area of cancer
pathogenesis and the use of bioactive sphingolipids in cancer treatment (Stover and Kester 2003;
8
Hannun and Obeid 2008; Simon et al. 2010). In the past three decades, sphingolipid research has
been an extremely active area however, due to the complex nature of sphingolipid metabolism
only a small number of highly significant mediators and signalers have been the research focus
(Hannun and Obeid 2008).
Changes in cellular levels of sphingolipids have also been observed at different stages in
cancer pathogenesis (Ogretmen and Hannun 2004). Ceramide and sphingosine-1-phosphate,
particularly, have been identified for their roles in mediating apoptosis and cell proliferation,
respectively (Ogretmen and Hannun 2004). Decreased levels of total ceramides and/or specific
long chain ceramides have been noted in several types of cancer. Total ceramide content, when
compared to normal ovarian cells, is less in ovarian tumors (Ogretmen and Hannun 2004). A
decrease in ceramides with acyl chains of 16, 18 and 24 carbons was noted in human lung, head
and neck cancers (Ogretmen and Hannun 2004). When coupled with observations that
sphingosine-1-phosphate levels increase, this may indicate that ceramide metabolism is
accelerated in cancerous cells producing higher amounts of sphingosine-1-phosphate (Ogretmen
and Hannun 2004; Hannun and Obeid 2008).
Exogenous ceramides have been introduced into a variety of cancer treatment scenarios
with positive results. The application of liposomally delivered ceramides can increase the rate of
apoptosis in cancer cells in vivo (Shabbits and Mayer 2003; Stover and Kester 2003). Stover and
Kester (2003) observed the effect of liposomally delivered ceramides on MDA-MB-231 breast
adenocarcinoma cells. When compared to free ceramide, liposomally encapsulated ceramides
that were delivered as part of a liposome contributed to a larger decrease in cancer cell
proliferation (Stover and Kester 2003). Shabbits and Mayer (2003) observed a similar trend with
liposomally introduced ceramides in their investigations of MDA435/LCC6 human breast cancer
in J774 mouse macrophage cells.
2.3 Supramolecular Strucures of Amphiphillic Lipids
Driven by the hydrophobic effect, amphiphilic lipids can orient in a variety of
supramolecular structures in aqueous solution (Tanford 1980; Israelachvili 1991). In order to
minimize or eliminate hydrocarbon-water interactions, phospholipids orient themselves so that
the hydrophilic headgroups ‘shield’ the hydrophobic hydrocarbon tails from the aqueous solution
(Israelachvili 1991). Characteristics of the molecule such as the charge and size of the headgroup
9
and the length and degree of saturation of the hydrocarbon tails effect how the molecules pack
together (Israelachvili 1991).
The favoured structure of amphiphilic molecules is determined by: the optimal effective
area of the head group (a); the volume of the hydrocarbon chain(s) (v), which is/are fluid and
incompressible; and the maximum effective length of the chain (lc) (Israelachvili 1991).
Depending on these packing considerations, five possible liquid-crystal configurations are
possible: spherical micelles, cylindrical micelles, flexible bilayer vesicles (i.e. liposomes), planar
bilayers and inverted micelles (Israelachvili 1991). Novel structures are also capable of forming;
these include both cubic and hexagonal liquid crystal structures as well as biphasic lipid
structures (Gan, Han et al., 2009).
Israelachvili has shown that in order for amphiphilic molecules to assemble into spherical
vesicles, the surface area (ao) must be sufficiently large, while the hydrocarbon volume (v) must
be sufficiently small and the radius of the supramolecular structure (R) must not exceed the
critical length of the amphiphilic molecule (Israelachvili 1991).
Based on geometric
considerations, for a sphere of radius, R, the mean aggregation number is equal to:
(Equation 2.1)
this may be rearranged to:
(Equation 2.2)
hence, the geometric constraint for a spherical vesicle may be calculated by:
(Equation 2.3)
where R=lc. Israelachvili has determined the mean dynamic packing parameters for different
liquid crystal morphologies (Israelachvili, 1991; Figure 2.2). Amphiphilic lipids with two
hydrocarbon chains are expected to form bilayers and liposomes due to their large hydrocarbon
volume resulting in a geometric constraint in the range of 1/2 to 1 (Israelachvili 1991).
10
Figure 2.5 Critical packing parameters for amphiphillic molecules and the resulting critical
packing shapes and resulting liquid crystal structures. Adapted from Israelachvili (1991).
11
2.3.1 Liposomes
Liposomes were first characterized in the early 1960’s by Bangham as bilayer vesicles in
aqueous solutions as formed by phospholipids (Bangham and Horne 1964; Bangham et al. 1965;
Bangham 1980). The characterization of liposomes occurred at a time when Bangham and
colleagues were investigating universal membrane structures (Bangham 1980).
The structure of phospholipids creates compartments within the liposome bilayer that
may contain exclusively lipid or aqueous solutions. This characteristic makes liposomes a
versatile tool for the encapsulation of compounds. As early as the 1970’s, the use of liposomes as
drug delivery systems was investigated (Gregoriadis 1980; Chonn and Cullis 1995). Initially, the
limitations of liposomal drug delivery pertained to the circulation time of the drug. In early
1990’s the in vivo circulation time of liposomal drugs were improved via the addition of
polyethylene glycol (PEG) (Klibanov et al. 1990). Current commercially available and clinical
trial versions of liposomal drugs are PEG-liposomes (Gregoriadis 1980; Chonn and Cullis 1995).
2.3.1.1 Advantages of Liposomal Drug Delivery Systems
The major downfall of chemotherapeutic drugs used in the treatment of certain types of
cancer is their lack of specificity. The same mechanisms by which wide spectrum chemotherapy
drugs such as doxorubicin or daunorubicin induce cancer cell apoptosis for cancer cells also
initiate the same events in healthy cells, which leads to vast complications including decreased
immune activity and subsequent secondary infections, chronic pain and fatigue (Strohl 1991).
Doxorubicin, for example, is believed to work via intercalation; the insertion of the drug into the
DNA of the cell. This interruption of the DNA sequence adversely affects the cancer cell’s
ability to correctly translate DNA into proteins, which ultimately reduces the rate at which the
cell successfully divides (Strohl 1991). Although doxorubicin is effective at destroying cancer
cells, it cannot distinguish between cancer cells and healthy cells thus doxorubicin is also a
potent mutagen. As a result, a potential side effect of doxorubicin used for chemotherapy is the
induction of other forms of cancer (Strohl 1991). As a drug that exclusively destroys cancer cells
has yet to be developed, using liposomal drug delivery systems to increase the specificity of
existing drugs provides the potential for a more effective cancer treatment strategy.
Liposomes are used clinically for the delivery of both water and lipid soluble drugs
(Allen 1996; Drummond et al. 1999). Liposomes have been shown to increase the circulation
12
time of drugs when compared to the free drug, however, the standard for drug delivery liposomes
are those that are sterically stabilized by the addition of PEG (Drummond et al. 1999). This
increase in circulation time of PEG-liposomal drugs over the free drugs utilized in cancer
chemotherapy regimes allows the exploitation of leaky tumor vasculature and increased blood
flow to the tumor, collectively know as enhanced permeability and retention (EPR) effects.
Cancer cells grow rapidly and without regulation of cell size or shape. Irregularities in cellular
properties result also create irregularities in the vasculature resulting in large blood vessels that
allow larger molecules such as liposomes to permeate the interior of the tumor.
Angiogenesis is the process by which new blood vessels are created. Tumors induce
blood vessel growth by releasing biochemical triggers that result in the creation of new vessels
that supply blood exclusively to the tumor. (Huang et al. 1992; Yuan et al. 1995; McDonald and
Baluk 2002). This increased blood flow to cancer cells allows for rapid growth that results from
increased nutrition and gas exchange (Huang et al. 1992; Yuan et al. 1995; McDonald and Baluk
2002). Liposomal drug delivery systems may exploit the utility of increased blood flow which,
based on their prolonged circulation, would result in a higher probability of drug-tumor
interactions (Huang et al. 1992; Yuan et al. 1995; McDonald and Baluk 2002).
The nature of tumor structure also results in decreased efficacy of lymph drainage
resulting in liposomes being retained within the tumor (Huang et al. 1992; Yuan et al. 1995;
McDonald and Baluk 2002).
2.3.1.2 Preparation Methods
Depending on the research focus and ultimate application of the liposome a variety of
production methods may be used to achieve the desired result. Size, lamellarity and the
homogeneity of the liposomal suspension are all end goals that influence preparation choice. The
first method used to prepare liposomes is referred to as the Bangham method, gentle hydration
method or lipid film hydration (Bangham 1980; Tsumoto et al. 2009; Lesoin et al. 2011).
Phospholipids are distributed in a thin film on the surface of glassware by dissolving the
phospholipids in an organic solvent, like choloroform, and later evaporating the solvent
(Bangham 1980; Tsumoto et al. 2009; Lesoin et al. 2011). The addition of an aqueous solution
under conditions of vortexing or agitation results in the production of liposomes (Bangham 1980;
Tsumoto et al. 2009; Lesoin et al. 2011).
13
The gentle hydration method often serves as the first step in other liposome production
methods. The phospholipids are suspended in the manner previously described and can be
physically processed with membranes or homogenization equipment to create liposomal
suspensions with specific physical properties (Gregoriadis 1995; Tsumoto et al. 2009; Lesoin et
al. 2011). The majority of experimentally produced and studied liposomes are extruded through
membranes under pressure in order to create a homogenous size distribution. Liposomes with
diameters between 100-200 nm are the most desirable as there is evidence that this size is the
most physiologically active (Gregoriadis 1995; Drummond et al. 1999). This size range does not
induce an immune response yet the liposomes are of sufficient size to carry an appropriate
concentration of the desired drug (Drummond et al. 1999).
In an effort to eliminate organic solvents, and any potential harmful residues that may
result, from the preparation of liposomes for drug delivery, novel preparation methods have been
developed (Lesoin et al. 2011). These include the preparation of liposomes using supercritical
fluids (Lesoin et al. 2011). In this process a phospholipid-solvent solution is sprayed through
supercritical carbon dioxide which causes the rapid evaporation of the solvent (Lesoin et al.
2011). This causes the phospholipids to precipitate out as empty vesicles that can then be
rehydrated using a variety of aqueous solutions (Lesoin et al. 2011). In cases where liposomes
are not being prepared for use in food or drug systems, the gentle hydration method represents an
opportunity to investigate phospholipid self assembly (Tsumoto et al. 2009).
14
3. DEPENDENCE OF LIQUID CRYSTAL MORPHOLOGY ON PHOSPHOLIPID
HYDROCARBON LENGTH1
3.1 Abstract
The liquid crystal morphologies of symmetrical diacyl phosphatidylcholine liposomes
examined in this research study were found to be dependent on saturated hydrocarbon chain
length. Both powder x-ray diffraction and synchrotron mid-IR microscopy indicate that
phosphatidylcholines with short hydrocarbon tails (i.e. ten and twelve carbons) are more likely to
form unilamellar liposomes while those with long hydrocarbon tails (i.e. eighteen and twenty
carbons) are more likely to form multilamellar liposomes. Hydrocarbon chain lengths of fourteen
and sixteen represent a transitional zone between these two liquid crystal morphologies. The
FTIR spectra where a shoulder develops on the peak at wavenumber 1750cm-1 particularly
highlights the change in the packing of adjacent molecules in the transitional zone.
3.2 Objective and Hypothesis
The objective of this research study is to assess the structural effects on liquid crystal
supramolecular structures of phospholipids caused by changing the length of the hydrocarbon
chains.
It is hypothesized that liposome diameter will increase in a positive linear fashion with an
increase in hydrocarbon length.
3.3 Introduction
Amphiphilic molecules have the ability to associate into numerous structures in aqueous
solutions depending on molecular packing considerations. The ability to pack into
1
Reprinted from Colloids and Surfaces B: Biointerfaces, 87 (1), H. Rutherford, N.H.
Low, F. Borondics, T. Pederson, M.A. Rogers. Dependence of liquid morphology on
phospholipid hydrocarbon length, 116-121, Copyright (2011), with permission from Elsevier.
15
supramolecular structures is governed by the hydrophobic attraction at the hydrocarbon-water
interface. As the molecules associate, the hydrocarbon chains pack together while the polar
entity of the molecule remains in contact with the aqueous phase. Hydrophilic, ionic or steric
repulsion of the polar head groups can further alter the packing of the molecules (Israelachvili
1991).
These contrasting parameters are a result of the amphiphilic nature of the molecules and
dictate the supramolecular structure (Tanford 1980). The favoured structure of amphiphilic
molecules is determined by: the optimal effective area of the head group (a); the volume of the
hydrocarbon chain(s) (v), which is/are fluid and incompressible; and the maximum effective
length of the chain (lc) (Israelachvili 1991). Depending on these packing considerations, a variety
of supramolecular structures are possible (Israelachvili 1991). Liposomes are spherical
supramolecular structures of phospholipids composed of at least one phospholipid bilayer and
are the structures examined in this study (Israelachvili 1991).
Israelachvili has shown that in order for amphiphilic molecules to assemble into spherical
vesicles, the surface area (ao) must be sufficiently large, while the hydrocarbon volume (v) must
be sufficiently small and the radius of the supramolecular structure (R) must not exceed the
critical length of the amphiphilic molecule (Israelachvili 1991). Based on geometric
considerations, for a sphere of radius, R, the mean aggregation number is equal to Equation 2.1
this may be rearranged to Equation 2.2 hence, the geometric constraint for a spherical vesicle
may be calculated by Equation 2.3. Israelachvili has determined mean dynamic packing
parameters for different liquid crystal morphologies (Israelachvili, 1991). Amphiphilic lipids
with two hydrocarbon chains are expected to form bilayers and liposomes due to their large
hydrocarbon volume resulting in a v/aolc in the range of 1/2 to 1 (Israelachvili 1991).
From the earliest investigations of phospholipid liquid crystal structures, particularly
liposomes, multiple bilayers structures have been observed (Bangham 1980; Gregoriadis 1980).
Research in the area of liposomes quickly began to focus in the ability of these structures to
effectively encapsulate compounds for a variety of applications including drug delivery
(Bangham 1980; Gregoriadis 1980; Drummond et al. 1999). Undoubtedly, the most resource
efficient way to study liposomes for encapsulation purposes is to prepare homogeneous
suspensions of unilamellar liposomes, often this is accomplished using sonication, membrane
extrusion or microfluidization (Gregoriadis 1995; Drummond et al. 1999; Lapinski, Castro-
16
Forero et al. 2007). Particularly when conducting research on liposomes for drug delivery
applications, control of both size and structure is of great concern since these parameters will
effect the performance of the drug delivery system in vivo (Allen 1996; Drummond et al. 1999).
Hydrocarbon chain length has been shown to have a positive linear relationship with
bilayer thickness (Lewis and Engelman 1983). The influence of hydrocarbon length on liposome
size can be overcome with processing; liposomes of a minimal size can be formed of
phospholipids with differing hydrocarbon chain lengths (Israelachvili et al. 1977). The effect of
the hydrocarbon chain length has not only been shown to affect the morphology of the liquid
crystalline structure but also the fluidity of the bilayer (Israelachvili 1991). While the formula for
geometric packing parameter (Equation 2.3) can be used to predict supramolecular structures, it
does not allow for the prediction of the structure of multilamellar liposomes. While unilamellar
liposomes have been considered to have more utility than their multilamellar counterparts, there
is great value in investigating the preferential aggregation of phospholipids with differing
hydrocarbon chain lengths into supramolecular structures without the aid of external factors such
as homogenization, filtration or ultrasonics.
3.4 Materials and Methods
The following synthetic symmetrical diacyl phosphatidylcholines were obtained from
Avanti Polar Lipids (Alabaster, AL, USA) and used as received: 1,2-didecanoyl-sn-glycero-3phosphocholine (PC10), 1,2-dilauroyl-sn-glycero-3-phosphocholine (PC12), 1,2-dimyristoyl-snglycero-3-phosphocholine (PC14), 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (PC16), 1,2distearoyl-sn-glycero-3-phosphocholine
(PC18),
and
1,2-diarachidoyl-sn-glycero-3-
phosphocholine (PC20). Liposomes were formed using a dry film gentle hydration method
(Bangham et al. 1965; Tsumoto et al. 2009). Samples were prepared in triplicate, all
measurements were performed on each sample. The film was prepared by dissolving 0.1 mM of
phosphatidylcholine in 2 mL of chloroform (Sigma-Aldrich, St. Louis, MO, USA) in a 50 mL
round bottom flask. A Buchi rotary evaporator (Flawil, Switzerland) with the water bath at 55oC
was used to evaporate the chloroform under reduced pressure leaving a thin dry film of
phosphtidylcholine. Two milliliters of HEPES buffered saline (20 mM HEPES, 150 mM NaCl)
(Sigma-Aldrich, St. Louis, MO, USA) was added to the film and was mixed using the rotary
action of the Buchi rotary evaporator with the water bath temperature set above the transition
17
temperature of the particular phosphatidylcholine being used. The resulting liposome
suspensions were sonicated using a Branson 2510 bath sonicator (Danbury, CT, USA) at room
temperature for 10 minutes at 40 kHz (i.e. until no large aggregates were visible in the
suspension).
A Nikon Eclipse E400 light microscope (Nikon Instruments Inc., Melville, NY, USA)
equipped with a Nikon DS-Fi1 color camera (Nikon Instruments Inc., Melville, NY, USA) and a
40x lens and condenser (Nikon Instruments Inc., Melville, NY, USA) were used for light
microscopy. Images were captured at a resolution of 2560 by 1920 pixels using a Nikon DS-FiL
color camera. Images were analyzed using Adobe Photoshop Extended CS4.0 (Adobe Systems
Inc., San Jose, CA, USA).
Light scattering measurements (hydrodynamic diameter and surface area) were
performed using a Malvern Hydro 2000S Mastersizer (Malvern, Worcestershire, UK).
Rheological experiments were performed at 25oC with a Thermal Analysis AR-G2 rheometer
(New Castle, DE, USA) using a 2o 40 mm acrylic cone and plate geometry. Oscillatory
frequency sweep measurements were conducted at room temperature and a constant stress of 1
Pa through a range of frequencies from 1 to 100 rad/s to assess storage modulus (G!) and loss
modulus (G") in the viscoelastic region.
Ten randomly selected vesicles were analyzed using Photoshop; the initial measurement
of diameter was taken in pixels and converted to microns using a calibrated scale bar. This
number-mean diameter (D[1,0]) was converted to a volume-moment mean diameter (D[4,3]) for
comparison with the collected light scattering data.
Fourier transform infrared (FTIR) spectra were collected at the Canadian Light Source
(Saskatoon, SK, Canada) at the mid-IR spectromicroscopy beamline (01B1-1). The end station
was comprised of a Bruker Optics IFS66v/S interferometer coupled to a Hyperion 2000 IR
microscope (Bruker Optics, Billerica, MA, USA). A drop of sample was placed between two
CaF2 optical windows (25 mm diameter, 2 mm thick). Light is focused on the sample using a
15X magnification Schwarzschild condenser, collected by a 15X magnification Schwarzschild
objective with the aperture set to a spot size of 40 mm by 40 mm and detected by a liquid
nitrogen cooled narrowband mercury cadmium telluride (MCT) detector utilizing a 100 mm
sensing element.
18
A KBr-supported Ge multilayer beamsplitter was used to measure spectra in the midinfrared spectral region. Measurements were performed using OPUS 6.5 software (Bruker
Optics, Billerica, MA). The measured interferograms were an average of 512 scans and were
recorded by scanning the moving mirror at 40 kHz (in relation to the reference HeNe laser
wavelength of 632.8 nm). The wavelength range collected was 690 to 7899 cm-1 with a spectral
resolution of 4 cm-1. Single channel traces were obtained using the fast Fourier transform
algorithm, with no zero-filling, after applying a Blackman–Harris 3-Term apodization function.
A Rigaku Multiplex Powder X-ray Diffractometer (Rigaku, Japan) with a ! degree
divergence slit, ! degree scatter slit, and a 0.3 mm receiving slit, was set at 40kV and 44 mA to
determine the polymorphic form and long spacings of the phosphatidylcholine vesicles. Scans
were performed from 1 to 30 degrees 2 theta at 0.2 degree/min.
Linear regression for Table 3.1 and ANOVA performed with Graphpad Prism 5 (San
Diego, California, USA).
3.5 Results and Discussion
The effect of sonication time on liposome size was assessed prior to deciding the final
preparation method. Light scattering measurements of preliminary samples were taken every
minute from zero to ten minutes, followed by measurements at five minute intervals up to 30
minutes. Ten minutes was sufficient to break apart large visible aggregates in the suspension, and
after ten minutes no significant changes in the size of the vesicles was noted. The application of
heat to levels above the transition temperature of the phosphatidylcholines during the mechanical
agitation of the suspension ensured that the entire lipid film was removed from the glass surface
and suspended in aqueous solution.
Regardless of hydrocarbon chain length, phosphatidylcholines from PC 10 to PC 20,
were observed to form spherical aggregates (Figure 3.1). Visualization, using light microscopy,
of these spherical aggregates could not conclusively discern whether the vesicles were composed
of single or multiple bilayers. The lower hydrocarbon chain lengths, PC 10-14, appear to have
small spherical particles which have aggregated into clumps during sample preparation (Figures
3.1A-C). Figures 3.1D-F, representing phosphatidylcholines with hydrocarbon chain length
greater than 16 carbons, appeared to have a slightly different morphology namely larger particles
with greater spaces between individual particles. As a note, it is not surprising to see a cylindrical
19
and spherical structures, such as in Figure 3.1D, as it is possible that both structures can exist in
thermodynamic equilibrium (Israelachvili 1991).
Figure 3.1 Light micrographs of vesicle suspensions composed of PC10 (A), PC12 (B), PC14
(C), PC16 (D), PC18 (E), and PC20 (F). Scale bar indicates 5 µm.
Previous research has revealed a positive linear relationship between hydrocarbon chain
length in phospholipids and bilayer thickness (Lewis and Engelman 1983; Allen 1996).
Assuming that a thicker bilayer would require a larger radius of curvature, it was expected that
liposome size would increase with increasing hydrocarbon chain length. Two separate trends
within the vesicle diameter collected using light microscopy were observed: the first being a
plateau from PC 10 to PC 14 and a second higher plateau from PC 16 to PC 20 (Figure 3.2).
20
Figure 3.2 Volume-moment mean diameters of phosphatidylcholine vesicles calculated using
pixel measurements obtained from light micrographs. Data points represent mean values
± one standard deviation.
To further assess the vesicle size trend, light scattering was used to determine the size of
the vesicles in solution. The raw percent volume versus particle size (Figure 3.3) indicated that
for short hydrocarbon chains the distribution of vesicle size was similar for chains shorter than
C14 (Figures 3.3A-C). A bimodal distribution was observed which suggests that, in concordance
with the light micrographs, that there were small vesicles and aggregates of small vesicles.
Above PC 14 (Figures 3.3D-F) the distribution shifts to larger particle sizes.
The calculated D[4,3] determined using light scattering (Figure 3.4A) showed similar
trends to those values determined using light microscopy (Figure 3.2). A linear tread in vesicle
size was observed above PC 14. Below a hydrocarbon chain length of PC 14, no significant
effect (p<0.05) of chain length on vesicle size was observed.
21
Figure 3.3 Light scattering intensities of phosphatidylcholine vesicle suspensions with varying
hydrocarbon chain lengths (A) PC10, (B) PC12, (C) PC14, (D) PC16, (E) PC18 and (F)
PC20.
22
Figure 3.4 Volume-moment mean diameters and surface area of phosphatidylcholine vesicles
calculated from dynamic light scattering measurements. Data points represent mean
values ± one standard deviation.
The parabola-like relationship of surface area and hydrocarbon chain length indicates that
the same type of structure is not being formed by the different chain length phosphatidylcholines.
Particularly, the increase in surface area from PC 10 to PC 12 and the abrupt decrease at PC 14
indicates a change in the structural dynamics of the vesicle structure. This suggests that the
packing arrangement may be affected by the hydrocarbon chain length resulting in a transition
from a unilamellar to multilamellar liposomes.
X-ray powder diffraction data indicated that the phosopholipid head groups were
arranged with a short spacing observed at ~ 4.2Å (Figure 3.5). However, phospholipids with
longer hydrocarbon showed a slightly smaller head group spacing (d~4.1 Å) (Figure 3.5A)
compared to the shorter hydrocarbon side chains (d~4.3 Å) (Figure 3.5B). This is attributed to
the smaller vesicle, which in turn results in a greater surface curvature, resulting in an increase in
the short spacing.
23
Figure 3.5 Powder X-ray diffraction data for PC18 (A) and PC12 (B).
X-ray diffraction patterns confirm the transition from unilamellar to multilamellar
structure with an increase in hydrocarbon chain length. This is evident by the 001 peak at 65.03
Å in Figure 3.5A. This indicates that the liposome has a bilayer which is equivalent to two times
the length on a single phospholipid. Further, the presence of the higher order reflections at 33.9
Å (002), 22.6 Å (003), and 16.9 Å (004) and 10.9 Å (006) indicate a liposome with multiple
bilayers (Figure 3.5A). As well, there are doublets peaks observed at approximately 22 Å and
16 Å which suggest two species are present. Conversely, a single long spacing at 29.5 Å was
observed for the C12 phospholipid suggesting that the liposome consists of a single bilayer of
24
phospholipid. This is confirmed by the long spacing being equal to the length of the phospholipid
as well as the lack of higher order reflections (Figure 3.5B). Synchrotron mid-IR spectroscopy
reveals a change in the short spacing of the head groups at the low and high end of the
hydrocarbon lengths examined. The FTIR data indicates differences in both the head group and
hydrocarbon chain packing for saturated phosphatidylcholines (Figure 3.6).
Figure 3.6 FTIR spectra for liposomal suspensions of PC 10 (bottom spectrum) to PC 20 (top
spectrum) vesicle suspensions in water (A) and deuterated water (B).
25
The peak at wavenumber ~1750cm-1 represents the C=O stretching of the ester bond
where the hydrocarbon chain meets the head group (Jiang et al. 2005). The appearance of a
shoulder on the 1750 cm-1 peaks for PC 14 and PC 16, indicates that the head group is more
highly confined (i.e., have a greater amplitude of hydrogen bonding). This indicates a transition
in the structure of liposomes formed below a chain length of 14 and above a chain length of 16.
PC 14 and PC 16 can be considered to have chain lengths within a transitional zone. This result
supports the observation in X-ray diffraction patterns (Figure 3.5A) where the phospholipids
with long hydrocarbon chains have a smaller headgroup spacing (i.e. pack more closely).
Logically, the multilamellar liposomes have a larger radius than the unilamellar liposomes,
resulting in less torsion applied to the adjacent molecules. As a result, in a multilamellar
liposome configuration adjacent molecules are closer and interact more highly with each other.
The hydrogen bonding associated with water make it difficult to discern the CH2
(symmetric stretch ~2850 cm-1; antisymmetric stretch ~2920 cm-1) and CH3 (symmetric stretch
was not evident; antisymmetric stretch ~2957 cm-1) bands of the hydrocarbon chain therefore
deuterated water was used to eliminate the interference of this spectral region (Figure 3.6B)
(Jiang et al. 2005).
Jiang et al. observed the formation of phospholipid bilayers on a titanium dioxide surface
(2005). The system presented here differs, though the differences caused by the curvature of the
bilayers in a liposomal system is considered to be negligible. Jiang et al. observed that when
PC16 formed a monolayer the CH2 symmetric stretch was observed at 2851 cm-1, when the
phosopholid layer increased to 3 layers the band shifted to 2850 cm-1 and at 5 layers it band was
observed at 2849 cm-1 (Jiang et al. 2005).
Similar trends were observed for the CH2
antisymmetric stretch where 1 layer had a band at 2919 cm-1, at 3 layers 2919 cm-1 and at 5
layers it shifted to 2918 cm-1 (Jiang et al. 2005). Figure 3.6B indicates that above C14 there is a
shift of both the CH2 antisymmetric and symmetric stretches to lower wavenumbers
corresponding to those observed as the layer thickness went from 1 phospholipid to multiple
phospholipid layers. The shift in the CH2 bands, along with the powder x-ray diffraction, concur
and suggest, in this system, that there is a transition form unilamellar to multilamellar structures
above a hydrocarbon chain length of 14 carbons.
While the majority of liposomes in the PC 10 sample are unilamellar, it is likely that
within each suspension both unilamellar and multilamellar liposomes exist. Previous research
26
has shown that multilamellar liposomes are present when samples are prepared using the gentle
hydration method. Further investigation would be required to assess the composition of samples
based on lamellarity. For the purposes of this study, the trends discussed will be determined
using the mean lamellarity indicated by the both the X-ray diffraction data and FTIR.
Oscillatory measurements were used to assess the storage and loss modulus of the
phospholipid suspensions, G! and G" respectively (Figure 3.7).
Figure 3.7 Oscillatory frequency measurements within the linear visco-elastic region for PC10
(A), PC12 (B), PC14 (C), PC16 (D), PC18 (E) and PC20 (F).
From Steffe’s work with rheology in soft matter systems, the behaviours of the
liposomal suspensions behave as dilute solutions (Steffe 1996). The crossover of the G’ and G”
values at the middle of the frequency range indicates that the suspension is behaving more like a
27
solid as it is subjected to higher frequencies. PC16, PC18 and PC20 suspensions exhibit this
transition. This indicates that multilamellar structures exhibit more solid-like properties under
oscillatory stress than the lipoomes formed using PC10, PC12 and PC14. By using simple
scaling relationship established
by Steffe a transition occurs and the hydrocarbon length
increases where an order of magnitude decrease in the G’ scaling constant (a) and an increase in
the order of magnitude in G” (b) is observed (Table 3.1). This observation confirms that
regardless of the phospholipid utilized it has a solution behaviour as a dilute solution. Further,
phospholipids with side chains less than C16, form numerous, small diameter liposomes which
are more capable in setting up an elastic network compared to the few large multilayer liposomes
formed with long phospholipid side chains.
Table 3.1 Power law coefficients of the storage (G! =a#b) and loss modulus (G" =c#d) of
liposomal suspensions compared with those for typical materials as determined by Steffe
(1996).
Hydrocarbon
a
b
c
d
Length
(Pa sb)
(n.u.)
(Pa sd)
(n.u.)
10
1.12 x10-2
1.17
5.96 x 10-4
2.02
12
1.33 x 10-2
1.10
8.10 x 10-4
1.89
14
1.19 x 10
-2
1.17
3.69 x 10
-3
2.12
16
1.18 x 10-2
0.98
9.09 x 10-2
1.08
18
3.60 x 10-3
1.88
1.07 x 10-1
0.94
20
4.00 x 10-3
2.15
1.43 x 10-1
1.02
Dilute Solution*
2.80 x 10-3
1.66
1.18 x 10-2
9.34 x 10-1
Concentrated
Solution+
Gel#
16.26
0.84
27.78
5.20 x 10-1
5626
3.70 x 10-1
344.70
1.45 x 10-2
*
5% dextrin;+ 5% lambda carrageenan; # 1% agar (Steffe 1996)
3.6 Conclusions
Phospholipid supramolecular aggregates may be modified from unilamellar to
multilamellar liposomes by altering the carbon number of the hydrocarbon chain. There is a
transition where below a chain length of 14 unilamellar liposomes form and above a chain length
of 16 multilamellar liposomes form. Evidence from FTIR spectra particularly indicates that
28
chain lengths of 14 and 16 represent a transitional zone in the liquid crystal morphology of
phospholipid liposomes. Understanding and characterizing this structural transition may have
utility in investigation the aggregation of amphiphilic molecules.
3.7. Rationale Linking Research Studies
Due to the similarities in the structures of PC and SM, research on the pairing of these
two phospholipids in liposomal structures precedes this study (Webb et al. 1995; Semple et al.
2005). Investigations of SM containing liposomes have been conducted in efforts to stabilize the
resulting liposomes and to mimic lipid raft structures found in cell membranes (Webb et al.
1995; Semple et al. 2005). However, what remains unexamined in literature is the effect of the
addition of SM to a liposomal drug delivery system on the overall efficacy of the treatment. In an
effort to facilitate further research in this area, a structural investigation was undertaken to
determine to what degree a two component system comprised of PC and SM would have on the
packing of adjacent molecules and effect on the overall supramolecular structure of the
phospholipids were altered in this two component system when compared to pure systems of PC
and SM.
Sphingomyelin is obtainable from both plant and animal sources, e.g. canola, eggs and
milk. Milk and its products (e.g. cheese) are a good source of sphingolipids (Vesper et al.,
1999). A possible waste stream for sphingomyelin isolation and utilization as a structural
component of liposomes is buttermilk (Morin et al., 2007). Phospholipids can be extracted from
buttermilk by filtration methods or by supercritical fluid extraction and sphingomyelin can be
isolated from this phospholipid fraction by the use of high performance liquid chromatography
(Astaire et al., 2003).
29
4. STRUCTURE OF SPHINGOMYELIN-PHOSPHATIDLYCHOLINE LIPOSOMES
4.1 Abstract
Using the gentle hydration method, liposomes composed of different ratios of 1,2distearoyl-sn-glycero-3-phosphocholine
(PC)
and
N-(tricosanoyl)-sphing-4-enine-1-
phosphocholine (SM) were formed. The addition of SM to a phosphatidylcholine (PC) liposome
at ratios of 1PC:1SM, 2PC:1SM and 4PC:1SM showed slight variations in packing. These are
the result of the enhanced hydrogen bonding capability of the SM headgroups and differences in
hydrocarbon chain length. Optimal packing in this study was found to occur at a ratio of
4PC:1SM.
4.2 Objective and Hypothesis
The objective of this research study is to assess the structural effects on liquid crystal
supramolecular structures of the addition of SM to a PC liposome at different ratios.
It is hypothesized that liposome diameter and the packing of adjacent molecules in a PCSM two component system will be similar to a single component PC liposome system.
4.3 Introduction
Research over the past 20 years has revealed the importance of sphingolipids in cell
signaling (Hannun and Obeid 2008). A number of these compounds including, sphingomyelin
(SM), ceramides, sphingosine and sphingosine 1-phosphate, (Figure 2.3) have been shown to be
of significant importance in cell signaling pathways associated with senescence and apoptosis
(Ogretmen and Hannun 2004; Hannun and Obeid 2008). As this class of compounds is capable
of inducing apoptosis, this function may be exploited to increase the rate of cell death when used
in cancer treatment.
Cancer cells grow rapidly and without regulation of size or shape (Huang, Lee et al.
1992; Yuan, Dellian et al. 1995; McDonald and Baluk 2002). Irregularities in cellular properties
propagate into irregularities in the vasculature of the tumor resulting in leaky blood vessels
30
which allow larger molecules, such as liposomes, to permeate the interior of the tumor (Huang et
al. 1992; Yuan et al. 1995; McDonald and Baluk 2002). The nature of tumor structure also
results in decreased efficacy of lymph drainage resulting in liposomal retention within the tumor
(Huang et al. 1992; Yuan et al. 1995; McDonald and Baluk 2002). The increased blood flow that
results from angiogenesis at the tumor site, coupled with long circulating liposomes results in a
higher probability of drug-tumor interactions (Huang et al. 1992; Yuan et al. 1995; McDonald
and Baluk 2002). These factors are collectively referred to as enhanced permeability and
retention effects (Huang et al. 1992; Yuan et al. 1995; McDonald and Baluk 2002).
Research has shown that the application of liposomally encapsulated ceramides can
increase the rate of apoptosis in cancer cells in vivo (Shabbits and Mayer 2003; Stover and
Kester 2003). Stover and Kester (2003) observed the effect of liposomally delivered ceramides
on MDA-MB-231 breast adenocarcinoma cells. When compared to free ceramide, ceramides that
were delivered as part of a liposome contributed to a larger decrease in cancer cell proliferation
(Stover and Kester 2003). Shabbits and Mayer (2003) observed a similar trend with liposomally
introduced ceramides in their investigations of MDA435/LCC6 human breast cancer in J774
mouse macrophage cells. Also, increased levels of dietary SM have been shown to decrease
cancer lesion size and delay cancer progression in MCF10AT1 breast cancer xenografts in mice
(Simon et al. 2010)
Sphingolipids are found in the membranes of cells where they are susceptible to
enzymatic degradation. A variety of enzymes have been shown to release these lipids from
cellular membranes into the cytoplasm or the extracellular space.
Once released these
compounds can participate in cell signaling cascades or synthetic pathways (Ogretmen and
Hannun 2004; Hannun and Obeid 2008). For example, the action of sphingomyelinase cleaves
the phosphatidylcholine head group from SM to produce ceramides. Released ceramides can
then stimulate a series of enzymes that carry out their own specific cellular effects leading to
senescence or apoptosis (Figure 4.1) (Ogretmen and Hannun 2004; Hannun and Obeid 2008).
31
Figure 4.1 Schematic representation of sphingolipid metabolism and their signaling activities,
adapted from Vesper et al. (1999).
Previous studies have revealed that the addition of sphingomyelin to a liposome structure
increases its stability (Webb et al. 1995; Semple et al. 2005). Increased stability is thought to
have been a result of resistance to enzymatic decay provided by the non-ester linkage of one of
the hydrocarbon chains to the SM head group as well as the high levels of saturation of the
hydrocarbon chains liposome (Webb et al. 1995; Semple et al. 2005). The more stable the
liposome structure and the more resistance to enzymatic decay, the longer the liposome can
circulate increasing the likelihood of the drug arriving at the disease site (Chonn and Cullis 1995;
Drummond et al. 1999).
It is proposed that a drug delivery liposome that contains sphingolipids, namely
sphingomyelin, has the potential to create a drug delivery system that can combine the effect of
liposomally encapsulated anti-cancer drugs with the anti-cancer effects of the released
sphingolipids. To our knowledge, this synergistic concept of the drug and bioactives within the
drug delivery system has yet to have been explored. Ideally, sphingomyelinase would act on the
32
liposome breaking apart the membrane resulting in (1) the release of the drug and (2) the release
of apoptosis-inducing ceramides. Sphingomyelinase has been shown to have activity on SM in
liposomal membranes, particularly in liposomes with lipid raft structures (i.e. lateral domains
composed of SM, PC and cholesterol) designed to mimic those found within mammalian cell
membranes (Nurminen et al. 2002). The rationale of conducting this research on sphingomyelinphosphatidlycholine liposomes is to determine whether or not there are any structural
impediments to an optimal liposome structure.
While both PC and SM possess a phosphocholine head group and two hydrocarbon
chains, there are differences in their structure that in turn effect their behaviour in phospholipid
bilayer systems (Terova et al. 2004). Phosphatidylcholine is synthesized from a glycerol
backbone to which the hydrocarbon chains are attached with ester linkages whereas SM is
composed of a sphingoid base to which an additional hydrocarbon chain is attached with an
amide bond (Terova et al. 2004). The presence of the amine group in the head group structure of
SM provides an additional site for hydrogen bonding that is not present on PC molecules (Terova
et al. 2004).
Sphingomyelin can be isolated from a variety of plant and animal sources. Depending on
the source, the hydrocarbon chain length and their saturation may differ. For example, milk SM
is more likely to have longer and less saturated hydrocarbon chains than SM extracted from egg
yolk (Astaire et al 2003).
4.4 Materials and Methods
The following were obtained from Avanti Polar Lipids (Alabaster, AL, USA) and used as
recieved: 1,2-distearoyl-sn-glycero-3-phosphocholine (PC18) and N-(tricosanoyl)-sphing-4enine-1-phosphocholine (SM) derived from milk. The predominant hydrocarbon species in milk
SM is 23:0 (Avanti Polar Lipids, Alabaster, AL, USA).
Liposomes were formed using a gentle hydration method (Tsumoto et al. 2009). The
following were prepared in triplicate: PC18, SM and samples with a molar ratio of 1PC:1SM,
2PC:1SM and 4PC:1SM. The film was prepared by dissolving 0.1 mM of total phospholipids in
2mL of chloroform (Sigma-Aldrich, St. Louis, MO, USA) in a 50 mL round bottom flask. A
Buchi rotary evaporator (Flawil, Switzerland) with the water bath at 55oC was used to evaporate
the chloroform under reduced pressure leaving a thin dry film of phosphtidylcholine. Two
33
milliliters of HEPES buffered saline (20 mM HEPES, 150 mM NaCl) (Sigma-Aldrich, St. Louis,
MO, USA) was added to the film and was mixed using the rotary action of the Buchi rotary
evaporator with the water bath temperature set above the transition temperature of the
phopholipid with the high transition temperature. The resulting liposome suspensions were
sonicated using a Branson 2510 bath sonicator (Danbury, CT, USA) at room temperature for 10
minutes at 40 kHz (until no large aggregates were visible in the suspension).
A Nikon Eclipse E400 light microscope (Nikon Instruments Inc., Melville, NY, USA)
equipped with a Nikon DS-Fi1 color camera (Nikon Instruments Inc., Melville, NY, USA) and a
40x lens and condenser (Nikon Instruments Inc., Melville, NY, USA) were used for light
microscopy. Images were captured at a resolution of 1920 by 960 pixels using a Nikon DS-FiL
color camera. Images were analyzed using Adobe Photoshop Extended CS4.0 (Adobe Systems
Incorporate, San Jose, CA, USA). ANOVA performed with Graphpad Prism 5 (San Diego,
California, USA).
Light scattering measurements were performed using a Malvern Hydro 2000S
Mastersizer (Malvern, Worcestershire, UK) to determine hydrodynamic diameter.
Fourier transform infrared (FTIR) spectra were collected at the Canadian Light Source
(Saskatoon, SK, Canada) at the mid-IR spectromicroscopy beamline (01B1-1). The end station
was comprised of a Bruker Optics IFS66v/S interferometer coupled to a Hyperion 2000 IR
microscope (Bruker Optics, Billerica, MA, USA). A drop of sample was placed between two
CaF2 optical windows (25 mm diameter, 2 mm thick). Light is focused on the sample using a
15X magnification Schwarzschild condenser, collected by a 15X magnification Schwarzschild
objective with the aperture set to a spot size of 40 mm by 40 mm and detected by a liquid
nitrogen cooled narrowband MCT detector utilizing a 100 mm sensing element.
A KBr-supported Ge multilayer beamsplitter was used to measure spectra in the midinfrared spectral region. Measurements were performed using OPUS 6.5 software (Bruker
Optics, Billerica, MA). The measured interferograms were an average of 512 scans and were
recorded by scanning the moving mirror at 40 kHz (in relation to the reference HeNe laser
wavelength of 632.8 nm). The wavelength range collected was 690 to 7899 cm-1 with a spectral
resolution of 4 cm-1. Single channel traces were obtained using the fast Fourier transform
algorithm, without any zero-filling, after applying a Blackman–Harris 3-Term apodization
function.
34
4.5 Results and Discussion
Visualization, using brightfield microscopy, reveals that the single component
phospholipid and sphingomylein systems (Figure 4.2) are composed mainly of spherical vesicles
with some cylindrical vesicles.
Figure 4.2 Light micrographs of liposomal suspensions composed of PC18 (A), SM (B),
1PC:1SM (C), 2PC:1SM (D) and 4PC:1SM (E) (Scale bar = 5 µm).
35
Israelachvili (1991) describes three molecular characteristics to calculate a packing
parameter that can be used to predict supramolecular structures. These characteristics are the
optimal area of the head group (ao); the volume of the hydrocarbon chain(s) (v), which are fluid
and incompressible; and the maximum effective length of the chain (lc) (Israelachvili 1991). The
packing parameter is calculated using v/aolc (Israelachvili 1991). The packing parameter can be
approximated by using the head group area for egg lecithin as determined by Israelachvili,
0.717nm2 (1991). Hydrocarbon volume (v) and length (lc) may be approximated using the
following equations:
v~(27.4 + 26.9n) x 10-3 nm3
(4.1)
lc ~ (0.154 + 0.1265n) nm
(4.2)
and
where n is the number of carbons in the aliphatic chain (Israelachvili 1991). Thus, the PC18
system has a packing parameter of 0.57 and the SM system, using also the approximate head
group area of 0.717nm2 and hydrocarbon chain length of 23:0, has a packing parameter of 0.36.
Packing parameters between 1/2 to 1 are expected to form liposomes while cylindrical micelles
have an expected packing parameter between 1/3 to 1/2 (Israelachvili 1991). The presence of the
cylindrical vesicles in the SM-PC samples is due to both the greater effective headgroup areas
and the increase in the hydrocarbon chain length of the SM molecules (Israelachvili 1991). In a
one component system, the hydrocarbon chains can pack uniformly which minimizes the effect
of hydrocarbon chain length on the supramolecular structure allowing for a more
thermodynamically stable structure (spherical vesicle formation) (Israelachvili 1991).
Figures 4.2A and 4.2E appear to have a similar distribution of liposomes and cylindrical
micelles. In the 4PC:1SM suspension, the majority of side-by-side interactions in this suspension
are PC-PC making it the most similar to the one component PC suspension. By minimizing
interactions between SM and PC the most optimal packing conditions of those examined were
observed for this two component system. If required, these morphological differences can be
readily overcome with physical modifications, such as membrane extrusion, which can be used
to prepare homogeneous size and shape distributions of liposomes regardless of phospholipid
composition or hydrocarbon chain length (Israelachvili et al. 1977).
Ten randomly selected vesicles from each light micrograph were analyzed using
Photoshop; the initial measurement of diameter was taken in pixels and converted to microns
36
using a calibrated scale bar. This number-mean diameter (D[1,0]) was converted to a volumemoment mean diameter (D[4,3]) (Figure 4.3). The volume-moment mean diameter shows no
significant difference (p < 0.05) in the single component SM liposomes and the SM-PC
liposomes. However, the SM-containing liposomes are smaller than the PC18 liposomes.
Figure 4.3 Volume-moment mean diameters of liposomal suspensions calculated using pixel
measurements obtained from light micrographs.
Light scattering showed a bimodal distribution for each suspensions which, in agreement
with the light micrographs, shows that both small vesicles and aggregates of small vesicles were
present (Figure 4.4). The size distribution pattern is similar for all suspensions. PC18 (Figure
4.4A) has a greater volume of vesicles between 0.1 and 1 micron and a less aggregates 10 micron
in size than the SM and SM containing suspensions (Figures 4.4B-E). The D[4,3] values
determined using DLS (Figure 4.5) shows similar trends as the values determined using
brightfield microscopy (Figure 4.3) for the SM-PC suspensions.
37
Figure 4.4 Light scattering intensities of liposomal suspensions of PC18 (A), SM (B), 1PC:1SM
(C), 2PC:1SM (D) and 4PC:1SM (E).
38
Figure 4.5 Volume-moment mean diameters of liposomal suspensions calculated from dynamic
light scattering measurements. Data points represent mean values ± one standard
deviation.
The difference in magnitude between the D[4,3] values, determined using microscopy
and DLS, is attributed to the light scattering technique detecting aggregates as opposed to single
vesicles. Figures 4.3 and 4.5 show opposing trends for the one component liposome systems. By
referring to Figure 4.2, however, these trends can be explained by the greater tendency of the SM
liposomes to aggregate (Figures 4.2B) due to the increased hydrogen bonding affinity of SM
headgroups (Niemela et al. 2004). From Figure 4.2, it is also observed that there are some minor
differences in the supramolecular structures in the liposomal suspension leading to some
discrepancies in the size measurements obtained from different techniques.
The previous
analyses have indicated few differences in the supramolecular structures of the prepared
suspensions. Further analysis using synchrotron mid-IR microscopy was conducted to assess the
impact on the physical chemistry of the systems. The FT-IR data indicates only small differences
in both the headgroup and hydrocarbon chain packing for all suspensions (Figure 4.6).
39
Figure 4.6 FTIR spectra for suspensions composed of PC, SM, 1PC:1SM, 2PC:1SM and
4PC:1SM.
The peak at ~1750 cm-1 represents the C=O stretch of the ester bond where the fatty acid
chain meets the head group (Jiang, et al. 2005). The single component SM liposomes displayed a
greater peak absorbance at this frequency than the single component PC liposomes (Figure 4.6).
This result suggests more that the head groups of SM are more highly confined due to their
increased capacity for hydrogen bonding.
40
The CH2 symmetric stretch (~2850 cm-1) and antisymmetric stretch (~2920 cm-1) and
CH3 antisymmetric stretch (~2957 cm-1) remained similar for all SM-containing liposomal
suspensions (Figure 4.6). The SM suspension show a slight increase intensity of this spectral
feature due to the introduction of longer, highly saturated hydrocarbon chains.
At a ratio of 1SM:4PC, the head group behaviour is similar to one component SM
suspensions
while ratios of 1SM:1PC and 1SM:2PC behave in a similar manner to one
component PC suspensions (Figure 4.6). Phosphatidylcholine has been used for liposomal drug
studies since the advent of research in this area and at present it is incorporated into drugs that
are approved for clinical usage making it an important benchmark to measure against.
Of the suspensions investigated, 1SM:4PC, seems to display the most optimal structure of
the systems tested. At this ratio, head group confinement is high despite the majority of
molecules being PC. Which, as previously described, is likely a result of ordered areas of PC
surrounding SM molecules minimizing interactions between unlike adjacent molecules. In all of
the SM-PC mixtures the level of head group confinement is never less than pure PC, indicating
that these two phospholipids in combination are capable of forming supramolecular structures
that are more stable than pure PC.
4.6 Conclusion
Sphingomyelin can be incorporated into phosphatidylcholine vesicles at numerous
possible ratios while maintaining size profiles and supramolecular structures similar to those
composed of only phosphatidylcholine. This versatility allows for possible modifications to drug
delivery systems so as to improve both their function and circulation. The incorporation of
sphingomyelin into a liposome that may be loaded with wide spectrum chemotherapy drugs
increases the utility of these drug carrier systems. The presence of sphingolipids, like
sphingomyelin, in the liposome structure may improve efficacy of current liposomally
encapsulated chemotherapeutics due to the possible activation of signalling pathways in which
sphingolipids induce apoptosis in cancer cell lines.
41
5. GENERAL DISCUSSION
Through the use of the gentle hydration method, spherical aggregates formed in all
suspensions investigated (Figures 3.1 and 4.2). The presence of particles with diameters
exceeding twice the length of the amphiphilic lipids indicated that rather than the formation of a
micelle, bilayer vesicles (liposomes) had formed. In order for micelles to form in the aqueous
system investigated, the hydrocarbon chains of amphiphilic lipids would be required to associate
at the centre of a sphere excluding any portion of the aqueous solution from within the structure.
The large aggregate size indicated that the inner vesicle contained aqueous solution. In addition
to spherical particles, cylindrical bilayer structures were also observed (Figures 3.1D and 4.2BD). The coexistance of these supramolecular structures is thermodynamically permissible in
suspensions.
Light scattering was used to observe trends in the diameter of the liposomal suspensions.
The size distributions of the prepared suspensions can be seen in Figures 3.3 and 4.4. Apart from
the suspension of PC 16 (Figure 3.3D), the predominant trend was a bimodal distribution size
(Figures 3.3 and 4.4). This bimodal distribution illustrated the capacity of the liposomes
investigated to aggregate in suspension. The SM containing suspensions have a distribution that
was skewed to higher particle sizes due to the increased hydrogen bonding affinity of SM
headgroups (Niemela et al. 2004). To investigate the lower end of the size distribution revealed
by light scattering, measurements of individual particles from the light micrographs were
examined (Figures 3.2 and 4.3). A comparison of Figures 3.2 and 3.4A which were both PC
suspensions, revealed a similar trend. The size of individual liposomes and the size of aggregates
of liposomes in the PC suspenions follow the same trend, whereas the SM liposome aggregates
are smaller than expected based on the individal SM liposomes. This is due to the enhanced
hydrogen bonding ability of SM. The difference in magnitude between the D[4,3] values
determined using microscopy and light scattering was attributed to the light scattering technique
42
used which detects aggregates as opposed to single vesicles, and the increased capacity of SM
containing liposomes to aggregate based on their increased hydrogen bonding capabilities.
The hydrocarbon chain length increase of saturated diacyl PC molecules did not result in
the expected linear increase in the diameter of the liposomes. The hypothesized outcome was
based on previous research showing a linear increase in membrane thickness with increasing
hydrocarbon length in PC molecules (Lewis and Engleman 1983). What was observed from both
pixel and light scattering measurements was a plateau from PC 10 to 14 and an increase above
PC 16. Measurements of surface area showed a parabola-like trend. Since all suspensions were
prepared with the same molar ratio these results indicated a change in the packing of PC
molecules above a chain length of 14 carbons (Figure 3.4B). The decrease in surface area
observed for PC 16-20 indicated a possible transition to multilamellar liposomes. The light
scattering data revealed trends that need to be further investigated with techniques appropriate
for the elucidation of the molecular structure of the liposomes.
To further investigate the makeup of the bilayers within the liposome, x-ray diffraction
was used. The 001 peak at 65.03 Å, which is twice the length of one PC 18 molecule, in Figure
3.5A indicated the presence of a bilayer. The higher order reflections (002-006) present indicated
the presence of multiple bilayers in Figure 3.5A and were absent from Figure 3.5B, which
represents PC 12. The single long spacing at 29.5 Å, which is twice the length of one PC 12
molecule as seen in Figure 3.5B, represents a suspension with a unilamellar liposome
composition. The existence of both unilamellar and multilamellar liposomes within the PC 18
suspension were suggested by the doublet peaks at ~22 Å and ~16 Å indicating that two different
structures were present.
Synchrotron mid-IR spectroscopy was used to elucidate the packing of adjacent
molecules in all experimental liposomal suspensions. The peak at wavenumber ~1750 cm-1
represents the C=O stretching of the ester bond where the hydrocarbon chain meets the head
group (Jiang et al. 2005). From Figure 3.6, the shoulder on the 1750 cm-1 peak for liposomal
suspensions of PC 14 and PC 16, indicated that the head group was more highly confined
(greater amplitude of hydrogen bonding) than the other suspensions. This observation of a
change in packing supports the x-ray diffraction data which showed a transition from unilamellar
to multilamellar lipsomes. In addition, a greater absorbance at this wavenumber was observed for
component SM versus single component PC liposomes (Figures 3.6 and 4.6). These results
43
indicated a higher confinement of SM head group structures due to their higher capacity for
hydrogen bonding (Niemela et al. 2004). Most similar to the pure SM suspension was the
suspension prepared at a ratio of 1SM:4PC. Ratios of 1SM:1PC and 1SM:2PC behaved in a
similar manner to one component PC suspensions (Figures 3.6 and 4.6).
The FTIR spectral features that correspond with the hydrocarbon chains of the
amphiphilic lipids are the CH2 symmetric stretch (~2850 cm-1) and antisymmetric stretch (~2920
cm1) and the CH3 antisymmetric stretch (~2957 cm-1). Jiang et al. (2005) observed the formation
of multiple bilayers of PC on a titanium oxide surface. When PC16 formed a monolayer the CH2
symmetric stretch was observed at 2851 cm-1 and when this phosopholipid layer increased to
three the band shifted to 2850 cm-1 ,and at five it shifted to 2849 cm-1 (Jiang et al. 2005). A
similar trend were observed for the CH2 antisymmetric stretch at 2919 cm-1 , 2919 cm-1, and
2918 cm-1 for one, three and five layers, respectively (Jiang, Gamarnik et al. 2005). This trend
was observed in this research (Figure 3.6B) for suspensions produced with a hydrocarbon chain
length of 14 where both the CH2 antisymmetric and symmetric stretches shifted to lower
wavenumbers corresponding to the transition from unilamellar to multilamellar structures.
Whereas, the hydrocarbon chain spectral features for SM containing suspensions (Figure 4.6)
remained similar. Any slight increase in the sample spectral features at wavenumbers ~2850 cm1,
~2920 cm1 and ~2957 cm-1 were attributed to the introduction of longer, highly saturated
hydrocarbon chains.
Rheological investigations of PC suspensions with differing hydrocarbon chain lengths
were conducted to assess the implications of lamellarity on viscoelastic behaviour. From Figure
3.7, suspensions of PC16, PC18 and PC20 showed a crossover of G! and G" values at the middle
(~10 rad s-1) of the frequency range which indicated more solid-like behaviour at to higher
frequencies (Steffe 1996).
44
6. GENERAL CONCLUSIONS
Phospholipids, as amphiphillic molecules, are capable of associating to form a variety of
supramolecular structures in aqueous solutions. Phospholipid vesicles have been used as model
systems for biological membranes and have been researched extensively. Phospholipid
liposomes were first characterized in the early 1960’s by Bangham as bilayer vesicles in aqueous
solutions (Bangham et al. 1965; Bangham 1980). It was postulated that the liposomal structure
was the most thermodynamically stable state as characterized by the reduction of an interfacial
tension between the aqueous medium and the liposome (Bangham, Standish et al. 1965;
Bangham 1980). The ability of liposomes to encapsulate both water and lipid soluble compounds
made them ideal candidates for drug encapsulation research (Bangham 1980; Gregoriadis 1980;
Allen 1996). Medical and pharmaceutical research employs these preparation methods in order
to achieve homogenous suspensions of unilamellar liposomes that are best suited to drug
encapsulation and delivery research (Gregoriadis 1980; Allen 1996; Drummond et al. 1999).
However, the analysis of phospholipid vesicle structures formed by ‘gentle’ methods
affords the opportunity to observe trends in their assembly that cannot be made when these
structures are produced through processing methods (Tsumoto et al. 2009). The value in utilizing
a method, such as the gentle film hydration method, is in part its simplicity as well as the ability
to relate the results found when using other processing methods to improve the engineering of
phospholipid vesicles for a variety of applications in the future (Tsumoto et al. 2009).
An increase in the saturated hydrocarbon chain length of PC molecules from X to Y did
not result in a positive linear increase in the size of liposomes which was expected based on
previous research that showed a positive linear increase in membrane thickness for the same
hydrocarbon chain increase (Lewis and Engelman 1983). Experimental results from this research
showed that below a hydrocarbon chain length of 14, PC was observed to associate into
unilamellar liposomes, while above a hydrocarbon chain length of 14 multilamellar liposomes
formed. FTIR spectra indicated that this was a result of a change in the packing behaviour of the
45
PC molecules due to the hydrocarbon chain length. Despite the change in liquid crystal
morphology, each suspension behaved as a dilute solution. Results from this work showed that
hydrocarbon length had an effect on the supramolecular structure of phospholipids which adds to
the body of knowledge on the systematic design of liposomes and self assembly of
phospholipids.
The addition of SM to a PC liposome at ratios of 1PC:1SM, 2PC:1SM and 4PC:1SM
were investigated. At these ratios, the structures of the liposome were not significantly different
from that of a single component PC liposome. However, the most optimal packing of adjacent
molecules was achieved at a ratio of 4PC:1SM. At this ratio the majority of adjacent molecule
interactions are PC-PC while the enhanced hydrogen bonding capabilities of the SM head group
allow for high confinement and a more ordered structure. This investigation of PC-SM liposomes
showed that this system was comparable to single component PC liposomes in terms of structure.
Based on these results, there should not be an impediment to investigating this system as a
possible drug delivery system. This versatility in liposome structures composed of two
components allows for modifications to a drug delivery system for improved functionality and
circulation. Particularly, the incorporation of SM into a liposome that is loaded with wide
spectrum chemotherapy drugs is an area that should be investigated for the possible synergistic
effects of the apoptitic effect of sphingolipids and cancer treatment drugs.
46
7. FUTURE DIRECTIONS
Though there are many processing methods that can effectively produce liposomal
suspensions for a wide variety of applications, this does not discount the value of investigating
‘minimal’ liposomal processing methods. Further research into the self assembly of liposomes
with different phospholipid characteristics and components may prove that these gentle methods
are viable option to current physical processing methods. More likely, however, is the gentle
hydration method leading to greater insight into the self assembly of phospholipids.
The inclusion of ceramide, a sphingolipid that plays a role in apoptitic pathways, into
liposomal structures has been shown to have anti-cancer activity (Shabbits and Mayer 2003;
Stover and Kester 2003).
Investigations of sphingomyelin as a liposome component have
focused on the creation of lipid raft-like structures, improved stability and in drug delivery
systems that include cholesterol (Webb et al. 1995; Semple et al. 2005). It is recommended that
future work includes systematic comparisons of systems that utilize phospholipid liposomes and
sphingolipid containing liposomes with and without drugs to assess whether a synergistic effect
between the anti-cancer drug and bioactive lipids is possible.
47
8. REFERENCES
Allen, T. M. (1996). "Liposomal Drug Delivery." Current Opinion in Colloid & Interface
Science 1(5): 645-651.
Astaire, J.C., R. Ward, et al. (2003). "Concentration of Polar MFGM from Buttermilk by
Microfiltration and Supercritical Fluid Extraction. Journal of Dairy Science. 86: 22972307.
Bangham, A. D. (1980). Development of Liposome Concept. Liposomes in Biological Systems.
G. Gregoriadis and A. Allison, Hoboken, NJ, USA, John Wiley & Sons Ltd.
Bangham, A. D. and R. W. Horne (1964). "Negative Staining Of Phospholipids And Their
Structural Modification By-Surface Active Agents As Observed In Electron
Microscope." Journal of Molecular Biology 8(5): 660.
Bangham, A. D., M. M. Standish, et al. (1965). "Diffusion Of Univalent Ions Across Lamellae
Of Swollen Phospholipids." Journal of Molecular Biology 13(1): 238.
Chonn, A. and P. R. Cullis (1995). "Recent Advances In Liposomal Drug-Delivery Systems."
Current Opinion in Biotechnology 6(6): 698-708.
Drummond, D. C., O. Meyer, et al. (1999). "Optimizing Liposomes For Delivery Of
Chemotherapeutic Agents To Solid Tumors." Pharmacological Reviews 51(4): 691-743.
Gan, L., S. Han, et al. (2010). "Self-assembled Liquid Crystalline Nanoparticles as a Novel
Ophthalmic Delivery System for Dexamethasone: Improving Preocular Retention and
Ocular Bioavailability." International Journal of Pharmaceutics. 396(1-2):179-187.
Goff, H.D. (1995). "University of Guelph Dairy Science and Technology Education." Retrieved
23 November 2009 from: http://www.foodsci.uoguelph.ca/dairyedu/home.html
Gregoriadis, G. (1980). The Liposome Drug-Carrier Concept. Liposomes in Biological Systems.
G. Gregoriadis and A. Allison, Hoboken, NJ, USA, John Wiley & Sons Ltd.
Gregoriadis, G. (1995). "Engineering Liposomes For Drug Delivery: Progress And Problems."
Trends in Biotechnology 13(12): 527-537.
48
Hannun, Y. A. and L. M. Obeid (2008). "Principles Of Bioactive Lipid Signalling: Lessons From
Sphingolipids." Nature Reviews Molecular Cell Biology 9(2): 139-150.
Hirabayashi, Y., Y. Igarashi, et al. (2006). Sphingolipid Synthesis, Transport and Cellular
Signaling. Sphingolipid Biology. Y. Hirabayashi, Y. Igarashi and A. H. Merrill, Springer.
Huang, S. K., K. D. Lee, et al. (1992). "Microscopic Localization Of Sterically Stabilized
Liposomes In Colon Carcinoma-Bearing Mice." Cancer Research 52(19): 5135-5143.
Israelachvili, J. N. (1991). "Aggregation Of Amphiphillic Molecules Into Micelles, Bilayers,
Vesicles And Biological Membranes." Intermolecular and Surface Forces. San Diego,
CA, USA, Academic Press.
Israelachvili, J. N., D. J. Mitchell, et al. (1977). "Theory Of Self-Assembly Of Lipid Bilayers
And Vesicles." Biochimica Et Biophysica Acta 470(2): 185-201.
Jiang, C. H., A. Gamarnik, et al. (2005). "Identification Of Lipid Aggregate Structures On Tio2
Surface Using Headgroup IR Bands." Journal of Physical Chemistry B 109(10): 45394544.
Johnston, M. J. W., S. C. Semple, et al. (2006). "Therapeutically Optimized Rates Of Drug
Release Can Be Achieved By Varying The Drug-To-Lipid Ratio In Liposomal
Vincristine Formulations." Biochimica Et Biophysica Acta-Biomembranes 1758(1): 5564.
Kent, C. (1995). "Eukaryotic Phospholipid Biosynthesis." Annuals Review of Biochemistry 64:
29.
Klibanov, A. L., K. Maruyama, et al. (1990). "Amphipathic Polyethyleneglycols Effectively
Prolong The Circulation Time Of Liposomes." Febs Letters 268(1): 235-237.
Lapinski, M. M., A. Castro-Forero, et al. (2007). "Comparison Of Liposomes Formed By
Sonication And Extrusion: Rotational And Translational Diffusion Of An Embedded
Chromophore." Langmuir 23(23): 11677-11683.
Lesoin, L., C. Crampon, et al. (2011). "Preparation Of Liposomes Using The Supercritical AntiSolvent (Sas) Process And Comparison With A Conventional Method." Journal of
Supercritical Fluids 57(2): 162-174.
Lewis, B. A. and D. M. Engelman (1983). "Lipid Bilayer Thickness Varies Linearly With Acyl
Chain-Length In Fluid Phosphatidylcholine Vesicles." Journal of Molecular Biology
166(2): 211-217.
Li, Z. and D. E. Vance (2008). "Phosphatidylcholine and Choline Homeostasis." Journal of Lipid
Research 49: 7.
49
MacGibbon, A.K.H. and M.W. Taylor. (2002). Phospholipids. In Encylopedia of Dairy Sciences
edited by Roginski, H., Fuquay, J.W. and P.F. Fox. San Diego, CA, USA, Academic
Press.
McDonald, D. M. and P. Baluk (2002). "Significance Of Blood Vessel Leakiness In Cancer."
Cancer Research 62(18): 5381-5385.
Morin, P., M. Britten, et al. (2007). "Microfiltration of Buttermilk and Washed Cream
Buttermilk for Concentration of Milk Fat Globule Membrane Components." Journal of
Dairy Science. 90: 2132-2140.
Niemela, P., M. T. Hyvonen, et al. (2004). "Structure And Dynamics Of Sphingomyelin Bilayer:
Insight Gained Through Systematic Comparison To Phosphatidylcholine." Biophysical
Journal 87(5): 2976-2989.
Nurminen, T. A., J. M. Holopainen, et al. (2002). "Observation Of Topical Catalysis By
Sphingomyelinase Coupled To Microspheres." Journal of the American Chemical
Society 124(41): 12129-12134.
Ogretmen, B. and Y. A. Hannun (2004). "Biologically Active Sphingolipids In Cancer
Pathogenesis And Treatment." Nature Reviews Cancer 4(8): 604-616.
Semple, S. C., R. Leone, et al. (2005). "Optimization And Characterization Of A
Sphingomyelin/Cholesterol Liposome Formulation Of Vinorelbilne With Promising
Antitumor Activity." Journal of Pharmaceutical Sciences 94(5): 1024-1038.
Shabbits, J. A. and L. D. Mayer (2003). "Intracellular Delivery Of Ceramide Lipids Via
Liposomes Enhances Apoptosis In Vitro." Biochimica Et Biophysica ActaBiomembranes 1612(1): 98-106.
Shingfield, K.J., Chilliard, Y., et al. (2008). Trans Fatty Acids and Bioactive Lipids in Ruminant
Milk. In Bioactive Components of Milk edited by Bosze, Z. Amsterdam, NL, Springer.
Simon, K. W., L. Tait, et al. (2010). "Suppression Of Breast Xenograft Growth And Progression
In Nude Mice: Implications For The Use Of Orally Administered Sphingolipids As
Chemopreventive Agents Against Breast Cancer." Food & Function 1(1): 90-98.
Steffe, J. (1996). Viscoelasticity. Rheological Methods in Food Process Engineering. East
Lansing, MI, USA, Freeman Press.
Stover, T. and M. Kester (2003). "Liposomal Delivery Enhances Short-Chain Ceramide-Induced
Apoptosis Of Breast Cancer Cells." Journal of Pharmacology and Experimental
Therapeutics 307(2): 468-475.
50
Strohl, W. R., Dickens, M.L., Raijgarhia, V.B., Woo, A.J., Priestley, N.D. (1991).
Anthracyclines. Biotechnology of Antibiotics. New York, Marcel Dekker.
Tanford, C. (1980). The Hydrophobic Effect: Formation of Micelles and Biological Membranes.
New York, Wiley.
Terova, B., J. P. Slotte, et al. (2004). "Miscibility Of Acyl-Chain Defined Phosphatidylcholines
With N-Palmitoyl Sphingomyelin In Bilayer Membranes." Biochimica Et Biophysica
Acta-Biomembranes 1667(2): 182-189.
Tsumoto, K., H. Matsuo, et al. (2009). "Efficient Formation Of Giant Liposomes Through The
Gentle Hydration Of Phosphatidylcholine Films Doped With Sugar." Colloids and
Surfaces B-Biointerfaces 68(1): 98-105.
Vesper, H., E. Schmelz, et al. (1999). "Sphingolipids in Food and the Emerging Importance of
Sphingolipids to Nutrition." Journal of Nutrition 129(7): 1239-1250.
Webb, M. S., T. O. Harasym, et al. (1995). "Sphingomyelin-Cholesterol Liposomes Significantly
Enhance The Pharmacokinetic And Therapeutic Properties Of Vincristine In Murine And
Human Tumor-Models." British Journal of Cancer 72(4): 896-904.
Yuan, F., M. Dellian, et al. (1995). "Vascular-Permeability In A Human Tumor Xenograft Molecular-Size Dependence And Cutoff Size." Cancer Research 55(17): 3752-3756.
51